OBJECTIVE: The purpose of this study is to evaluate the long-term usage of the hypoglossal nerve stimulator and identify predictors of usage over time. STUDY DESIGN: Retrospective chart review. SETTING: Tertiary academic medical center and database. METHODS: Single institution retrospective chart review was merged with Inspire SleepSync device usage data. Growth curve modeling was performed with a linear mixed-effects model to evaluate usage hours over time. A logistic regression analysis was also performed to assess the relationship between early adherence and adherence at 1 and 2 years. RESULTS: A total of 352 patients were included in the study. Seventy percent were male and the mean age was 64 ± 12 years. Adherence at 6 months was significantly predictive of adherence at 12 (odds ratio [OR] 9.42, CI 3.23-28.76, P <
.001) and 24 (OR 3.03, CI 0.96-9.05, P = .049) months. Older age (estimate = 0.019, SE 0.006, P = .002) and greater days of use per biweekly period (estimate = 0.142, SE 0.004, P <
.001) were positive predictors of usage. Insomnia (estimate = -0.386, SE 0.128, P = .003) and higher device mean voltage amplitude (estimate = -0.348, SE 0.020, P <
.001) were negative predictors. CONCLUSION: Time since activation does not appear to have an impact on hypoglossal nerve stimulator therapy usage. Insomnia and higher device voltage were associated with a statistically significant negative trend in therapy usage, whereas older age and greater number of days used may be predictive of higher therapy usage. Identifying predictors of usage will enable identification of at-risk patients and allow for early implementation of targeted interventions to improve patient adherence and optimize long-term health outcomes.